Acta Chiropterologica, 6(2): 177–192, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Phylogeny of African Myotis (Chiroptera, ) inferred from cytochrome b sequences

BENOÎT STADELMANN1, 2, DAVID S. JACOBS3, CORRIE SCHOEMAN3, and MANUEL RUEDI1, 4

1Natural History Museum, P.O. Box 6434, 1211 Geneva 6, Switzerland 2Department of Zoology and Biology, Molecular Systematics Group, University of Geneva, 30 quai Ernest-Ansermet, CH-1211 Geneva 4, Switzerland 3Department of Zoology, University of Cape Town, Private Bag, Rondebosch 7701, South Africa 4Corresponding author: [email protected]

The genus Myotis is comprised of about 100 species that are unequally distributed between the Northern (81% of the species) and the Southern hemisphere (19% of the species). Only eight species of Myotis occur in the Ethiopian region, but this is the only biogeographic region with representatives of all four classical subgenera, suggesting a diverse assemblage of morphotypes. We used sequences of a mitochondrial DNA gene (cyt b) to investigate the evolution and the phylogenetic position of seven of the eight Ethiopian species, and compared them to a broad sampling of Myotis from the World and of other vespertilionids. Phylogenetic reconstruction was based on 91 complete sequences representing 79 species of bats. The two endemic southern African species of the subgenus Cistugo were not placed within the genus Myotis, but were basal to the vespertilionid radiation, as suggested by earlier work based on karyology. The remaining Ethiopian species formed a strong mono- phyletic clade within Myotis, further stressing the importance of biogeography as a good predictor of phyloge- netic relationships. This Ethiopian clade includes one Western Palaearctic and one Oriental species, both of which probably secondarily colonized these areas from the Ethiopian region. Molecular dating based on Bayesian inferences suggest that these faunal exchanges occurred at the end of the Miocene, while the split of the Ethiopian clade from the other Old World Myotis dates back to the middle Miocene, quite early in the Myotis radiation. Thus, the relative paucity of species in sub-Saharan Africa cannot be attributed to a late entry into this continent. Instead, these molecular results suggest that other evolutionary processes are responsible for the poor species diversity of Myotis found in Africa today.

Key words: Cistugo, African Myotis, Vespertilionidae, cytochrome b, molecular dating

INTRODUCTION (55 species), followed by the Nearctic re- gion (24 species), while only 14 species are With about 100 currently recognized found in the Neotropics, eight in the species (Koopman, 1994; Simmons, In Ethiopian, and three in the Australian region press), bats of the genus Myotis represent (Koopman, 1994; Horá…ek et al., 2000). one of the few mammalian groups with a This pattern is particularly surprising since natural distribution covering most of the Myotis bats have adapted to diverse habi- world. The peak diversity of Myotis species tats, from deserts to cloud forests, with a is, however, reached in the northern conti- wide range of foraging habits (from insecti- nents. Eurasia supports the richest diversity vorous aerial hawkers to fish-eating). They Acta Chiropterologica, 6(2): 193–217, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Systematic status of African populations of pipistrellus complex (Chiroptera: Vespertilionidae), with a description of a new species from Cyrenaica, Libya

PETR BENDA1, PAVEL HULVA2, and JIÌÍ GAISLER3

1Department of Zoology, National Museum (Natural History), Václavské nám. 68, 115 79 Praha 1, Czech Republic; E-mail: [email protected] 2Department of Zoology, Charles University, Vini…ná 7, 128 44 Praha 2, Czech Republic 3Department of Zoology and Ecology, Masaryk University, KotláÍská 2, 611 37 Brno, Czech Republic

The distribution of pipistrelles of the Pipistrellus pipistrellus complex (= P. pipistrellus s.l.) reaches only marginally the African continent. These bats are known only from a narrow belt of the Mediterranean zone in Maghreb and from NE Libya. We analysed museum specimens of African populations of P. pipistrellus s.l. using both morphologic and genetic techniques and compared them with Eurasian specimens of the complex. The African representatives of P. pipistrellus complex include two morphologically, genetically and geographically distinct populations. One of them inhabits the Mediterranean part of Cyrenaica, Libya. Belonging to the P. pygmaeus genetic lineage, these bats are represented by larger and more rusty coloured individuals with large massive rostrum and canines. In morphologic traits, this population differs significantly from all Western Palaearctic populations of the P. pipistrellus complex. These bats differ by about 6–7% in genetic distance from P. pygmaeus s. str. Within the P. pygmaeus lineage Libyan bats seem to be unique in their echolocation calls: the maximum energy of terminal frequencies was recorded at about 45 kHz. We consider the Libyan pipistrelles to represent a separate species, Pipistrellus hanaki sp. nov. Another distinct African pipistrelle population inhabits the Mediterranean parts of NW African countries, Morocco, Algeria and Tunisia. Individuals from the latter population are small and somewhat darker members of the P. pipistrellus genetic lineage, with relatively short and narrow mesial part of rostrum. Although both morphological and genetic differences between this population and Eurasian P. pipistrellus s. str. were found (genetic distance about 3–5%), they are probably not sufficient for the separation of this form at the specific level. However, the differences from European samples show rather not a cline character and therefore potential subspecific level of NW African P. pipistrellus has to be taken into consideration.

Key words: Pipistrellus pipistrellus, Pipistrellus pygmaeus, Africa, morphology, morphometry, cytochrome b, mitochondrial DNA,

INTRODUCTION East up to Central Asia (Corbet, 1978; Koop- man, 1994; Taake and Vierhaus, 2004). Dis- Pipistrelle bats of the Pipistrellus pipi- tribution range reaches only the northern- strellus complex (or P. pipistrellus sensu la- most margin of the African continent, where to) are widely distributed in the western part these bats are known solely from the narrow of Eurasia, from Iberia, British Islands and belt of the Mediterranean zone in Maghreb southern Scandinavia through the Middle and from NE Libya (Fig. 1). Acta Chiropterologica, 6(2): 219–226, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

A new species of Kerivoula (Chiroptera: Vespertilionidae) from Myanmar (Burma)

PAUL J. J. BATES1, MATTHEW J. STRUEBIG2, STEPHEN J. ROSSITER2, TIGGA KINGSTON3, SAI SEIN LIN OO4, and KHIN MYA MYA 4

1Harrison Institute, Centre for Systematics and Biodiversity Research, Bowerwood House, St. Botolph’s Road, Sevenoaks, Kent, TN13 3AQ, United Kingdom; E-mail: [email protected] 2School of Biological Sciences, Queen Mary, University of London, London, E1 4NS, United Kingdom 3Department of Geography, Boston University, 675 Commonwealth Ave, Boston, MA 02215, USA 4Department of Zoology, University of Mandalay, Mandalay, Myanmar

A new species of Kerivoula is described. Currently, it is known only from Namdee Forest in southern Kachin State, Myanmar. Externally, superficially similar to Kerivoula papillosa and with a dentition comparable to that of Kerivoula lenis, it is distinguished by its flattened skull. It was collected in evergreen forest in an area that also includes some mixed deciduous forest, shifting cultivation, and bamboo groves.

Key words: Kerivoula sp. nov., Kachin State, Myanmar (Burma)

INTRODUCTION pers. comm.). This team, with its experience of working in forest habitats elsewhere in In November 1999, the Harrison Insti- South-east Asia, had a particular interest in tute and the University of Yangon conduct- the forest bats of northern Myanmar. ed the first international field survey for Results of their preliminary studies in bats in Myanmar for fifty-five years. Four southern Kachin showed a particularly rich years later, following the award of a Darwin fauna, with fourteen species collected Initiative grant, an international bat work- from two survey sites. At least one of these shop in the University of Yangon, numerous represents a previously unknown species of additional field surveys and 10 internation- Kerivoula, which is here described and al expeditions to 12 of the 14 states and di- compared to other taxa within the subfami- visions of Myanmar, the programme of bat ly Kerivoulinae, Asiatic representatives of studies continues to develop. Results have which were recently reviewed by Vanitha- been, and are being, published in a series of rani et al. (2003). scientific papers and reports (Bates et al., 2000, 2001, 2004; Pearch et al., 2003). MATERIALS AND METHODS In March 2003, a second team of UK scientists joined forces with students and Field Survey staff of the University of Mandalay to con- Namdee Forest was one of twenty-five localities duct additional researches (M. J. Struebig, in Kachin and Shan States and Mandalay and Sagaing Acta Chiropterologica, 6(2): 227–248, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

A new species of Chaerephon (Molossidae) from Madagascar with notes on other members of the family

STEVEN M. GOODMAN1 and SCOTT G. C ARDIFF2

1Field Museum of Natural History, 1400 South Lake Shore Drive, Chicago, Illinois 60605, USA, and WWF, BP 738, Antananarivo (101), Madagascar; E-mail: [email protected] 2Columbia University, Department of Ecology, Evolution, and Environmental Biology, 1200 Amsterdam Avenue, New York, NY 10027, and Division of Vertebrate Zoology (Mammalogy), American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, USA

We describe a species of Chaerephon (Molossidae) new to science from western Madagascar. This bat differs from the other two Chaerephon occurring on the island and from comparably sized African and Asian Chaerephon based on measurements, pelage and wing coloration, and cranial and dental characters. Chaerephon sp. nov. occurs at three sites in the drier western portion of the island. We also provide some natural history and distributional information on other Malagasy members of this family.

Key words: Chaerephon, Molossidae, new species, distributional records, western Madagascar

INTRODUCTION coastal and inland areas and it is also re- ported from the eastern offshore island of The bats of Madagascar are poorly Isle Sainte Marie (Peterson et al., 1995). It known. As currently understood, seven taxa can be found roosting in monospecific col- represent the Molossidae fauna of the is- onies or sharing sites with other molossid land: Mormopterus jugularis, Tadarida ful- bats such as M. jugularis and M. leuco- milans fulminans, Chaerephon leucogaster, stigma. Chaerephon pumilus occurs in areas C. pumilus, Mops leucostigma, M. midas of the humid east and central highlands and miarensis, and Otomops madagascariensis the vast majority of records are associated (Peterson et al., 1995; taxonomy and spell- with human habitations (Peterson et al., ing in accordance with Simmons, In press). 1995). Its range at least partially overlaps Given the difficulty of capturing members with M. jugularis (see below). of this family, few data are available about During the course of biological invento- their distribution and natural history, even ries carried out over the past decade at dif- for the common commensal species. ferent sites on the island (Fig. 1), particular- On Madagascar two species of Chae- ly the western portion, we have captured rephon are currently recognized: C. leu- a number of bats, including Molossidae. cogaster and C. pumilus. The former spe- Amongst these collections is a species of cies often occurs as a commensal, living in Chaerephon that is unknown to science. man-made shelters, along the drier western The purpose of this paper is to describe the Acta Chiropterologica, 6(2): 249–268, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Systematic notes on a collection of bats from Malawi. I. Megachiroptera: Epomophorinae and Rousettinae (Mammalia, Chiroptera)

WIM BERGMANS1 and NICO J. VAN STRIEN2

1Zoölogisch Museum, Universiteit van Amsterdam, P.O.Box 94766, Amsterdam, the Netherlands E-mail: [email protected] 2Julianaweg 2, 3941 DM Doorn, the Netherlands

From July 1986 to August 1991 the junior author collected bats in various localities in Malawi. About 450 of these were preserved. In the present study the Megachiroptera are reported. Epomophorinae (Megachiroptera) collected in Malawi by H. Jachmann in 1982 are also reviewed. In addition, a number of specimens in collections in Malawi, Zambia and Zimbabwe have been examined by the junior author, while data on some other samples from Malawi in various other collections, examined earlier by the senior author, have been reconsidered as well. The combined collections contain eight species of Megachiroptera, including Epomophorus cf. labiatus, E. gambianus crypturus, a new species of Epomophorus (described herein), E. wahlbergi, Epomops dobsonii, Rousettus aegyptiacus leachii, R. lanosus, and Eidolon helvum. For the new species, an IUCN Red List category is proposed.

Key words: Megachiroptera, Malawi, systematics, Epomophorinae, Rousettinae

INTRODUCTION (Bocage, 1889), Eidolon helvum (Kerr, 1792), Rousettus aegyptiacus (É. Geoffroy- Since the 1980s, Dr. D. C. D. Happold St. Hilaire, 1810), and R. lanosus Thomas, and Mrs. M. Happold have published sever- 1906. Kock et al. (1998) recorded the eighth al reports on the bats of Malawi, the first co- Megachiroptera, Plerotes anchietai (Seabra, authored by the late J. E. Hill (Happold et 1900). al., 1987, 1989, 1997). These studies are The objective of the present paper is to mainly faunistic, dealing with species in- contribute to the knowledge of the taxon- ventories, distribution, ecology, status and omy of Megachiroptera from Malawi. The conservation. Happold et al. (1997) listed taxonomic status of various forms of Epo- the 59 bat species known to occur in Ma- mophorus and Rousettus in Malawi is re- lawi, including seven species of Megachi- viewed, and a new species of Epomophorus roptera (they did not deal with subspecies): is described. The study builds on earlier Epomophorus gambianus (Ogilby, 1835), ones dealing with or including the same E. labiatus (Temminck, 1837), E. wahlber- subject (Bergmans et al., 1983; Bergmans, gi (Sundevall, 1846), Epomops dobsonii 1988, 1994). Bergmans (1988, 1994) gave Acta Chiropterologica, 6(2): 269–292, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Changes in bat fauna during the Middle and Late Holocene as exemplified by thanatocoenoses dated with 14C AMS from Kraków-Cz“stochowa Upland caves, Poland

TOMASZ POSTAWA

Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, ul. S»awkowska 17, 31–016 Kraków, Poland; E-mail: [email protected]

Fossil material from 12 caves situated in the Kraków-Cz“stochowa Upland (southern Poland) was studied. In 17 samples of osteological material, 8,275 individuals (Minimal Number of Individuals, MNI = 4,571) of 12 bat species were identified. Thirteen separate thanatocoenoses were found (R × C test) from which bone material was dated using the 14C AMS method; 13 dates from 6,725 to 820 14C yr BP (yr BP) were obtained. Correspondence analysis for thanatocoenoses from the Atlantic and Subboreal periods distinguished two bat species groups and two thanatocoenosis groups: 1) pontic-mediterranean species with a higher frequency during the Atlantic period, e.g., Rhinolophus hipposideros which reoccurs more abundantly at the end of the Subboreal period and Myotis emarginatus, which is absent in earlier sediments with the exception of one episode; 2) Myotis daubentonii, auritus, and Myotis dasycneme which increase in frequency during humid and cool periods (5,500–4,200 and 3,000–2,700 yr BP); 3) thanatocoenoses from the Holocene climatic optimum (6,000–5,500 yr BP) characterized by a dominance of Myotis nattereri; a decrease in its frequency occurred during the cooler end of the Atlantic period; 4) thanatocoenoses from the Subboreal period (4,100–3,500 yr BP) characterized by a dominance of Myotis bechsteinii. The maximum frequency of M. bechsteinii correlates with an increasing share of Fagus and Carpinus in forest ecosystems, while its decrease was probably caused by disease and was independent of human activity. Myotis myotis was found in assemblages from the Atlantic period, while the remains of a nursery colony in Nietoperzowa Cave (820 ± 25 yr BP) indicate that reproduction of this species occurred to the north of the Carpathians before the appearance of houses with attics. The presence of mass concentrations of Pipistrellus pipistrellus (s.l.) in caves was confirmed for the Subatlantic period (2,325 ± 30 yr BP), which shows its independence from both thermal balance and human influence on contemporary ecosystems. The low frequency of barbastellus in thanatocoenoses prevents reconstructions. Reconstructions for the Atlantic and Subboreal periods show that the composition of the bat fauna depends on changes in climate and vegetation, while human activity seems to have marginal impact. A comparison of paleozoological and radiocarbon datings revealed large differences in age estimation of the thanatocoenoses.

Key words: Kraków-Cz“stochowa Upland, Poland, Holocene, bat fauna, 14C AMS, cave, climate, vegetation change, human activity

INTRODUCTION beginning is dated from 10,250 14C yr BP in Poland and continues to this day (Starkel, After the Pleistocene Vistulian glacia- 1999). As the climate warmed to the north tion, the contemporary warm period of of the Carpathian and Sudetes Mountains, the Quaternary, Holocene, had begun. Its the contemporary flora and fauna began to Acta Chiropterologica, 6(2): 293–302, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Geographic distribution, ecology, and phylogenetic affinities of Thyroptera lavali Pine 1993

SERGIO SOLARI1, RONALD A. VAN DEN BUSSCHE2, STEVEN R. HOOFER1, and BRUCE D. PATTERSON3

1Department of Biological Sciences, Texas Tech University, Lubbock, TX 79409, USA E-mail: [email protected] 2Department of Zoology and Collection of Vertebrates, Oklahoma State University, Stillwater, OK 74078, USA 3Department of Zoology, Field Museum of Natural History, Chicago, IL 60605, USA

Thyroptera lavali (Chiroptera: Thyropteridae) is a rare Neotropical species that until now has been recorded from only five localities in the Amazonian rainforests of Peru, Ecuador, Venezuela, and Brazil. Fewer than 10 specimens of T. lavali exist and, accordingly, little is known about its distribution, natural history, and phylogenetic affinities. We report new records for the species from southeastern Peru. Together with other recently published records, these expand the known range of the species considerably, as well as increase our knowledge of its ecology. Thyroptera lavali seems to prefer primary forest near swamps, and probably roosts in palms; its reproductive pattern is similar to that of other Neotropical insectivorous bats, with parturition at the beginning of wet season. Finally, we used two different data matrices to assess its phylogenetic relationships: one of discrete morphological characters, the other of DNA sequences of mitochondrial genes. Both data sets support a sister relationship between T. lavali and T. tricolor, with T. discifera as the basal member of the genus Thyroptera.

Key words: disc-winged bats, distribution, ecology, Neotropics, systematics, Thyroptera lavali, Thyropteridae

INTRODUCTION Peru (Pine, 1993). Previous reports of T. lavali include a fossil record in Colombia The family Thyropteridae (disc-winged (as T. robusta from Magdalena Valley, bats) includes three species in the genus Czaplewski, 1996, 1997), a northern record Thyroptera that are endemic to lowland in Napo Province, Ecuador (Reid et al., Neotropical forests (mainly below 1,000 m; 2000), the type locality (Quebrada Espe- Koopman, 1978; Patterson et al., 1996). ranza, Loreto; Pine, 1993), and the first Thyroptera discifera and T. tricolor have extension southward (Bosque Von Hum- extensive distributions throughout the boldt, Ucayali; Solari et al., 1999). These Amazon basin and beyond (Wilson and records extended the latitudinal range more Findley, 1977; Wilson, 1978). The third than 600 km north and 650 km south of the species, T. lavali, was described on the basis type locality, clearly indicating that the of four specimens from a single lowland species has a more widespread distribution locality in Loreto department, northeastern than previously known. Two recent records Acta Chiropterologica, 6(2): 303–308, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Habitat use of the Pacific sheath-tailed bat (Emballonura semicaudata) on Aguiguan, Mariana Islands

JACOB A. ESSELSTYN1, 2, GARY J. WILES3, and ARJUN AMAR1, 4

1Division of Fish and Wildlife, Rota, Northern Mariana Islands 96951, USA 2Present address: Natural History Museum and Biodiversity Research Center & Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 66045, USA; E-mail: [email protected] 3Washington Department of Fish & Wildlife, 600 Capitol Way North, Olympia, WA 98501, USA 4Present address: Centre for Ecology and Hydrology-Banchory, Hill of Bratherns, Banchory, Aberdeenshire, AB31 4BW, United Kingdom

We tested for differential habitat use by Pacific sheath-tailed bats (Emballonura semicaudata) in three major vegetation types on Aguiguan, Mariana Islands. Acoustic surveys of bat activity were conducted on a 370-m grid with 50 stations that covered the entire island. We controlled for a variety of extraneous factors through sampling design and use of a generalized linear model. Bat activity was significantly higher in native and non- native forests than in non-forest habitats. There was no significant difference between activity levels in the two forest types. However, our sample size in non-native forest was limited to seven stations, thus the conclusion that non-native forest is an important habitat for the species should be viewed with caution. Our finding that E. semicaudata may be reliant on forest is critical because forests on Aguiguan are threatened by feral goats.

Key words: Emballonura semicaudata, Pacific sheath-tailed bat, habitat use, Aguiguan, Mariana Islands, bat detector, conservation

INTRODUCTION declines or extirpation (e.g., Marianas, Fiji, and the Samoas — Lemke, 1986; Grant et The Pacific sheath-tailed bat (Emballo- al., 1994; Flannery, 1995; Hutson et al., nura semicaudata) is broadly distributed 2001; Tarburton, 2002). Possible causes of across the tropical Pacific. The species is these declines have been proposed (e.g., ef- known from the Mariana and Caroline Is- fects of war and pesticides), but no hypoth- lands, Fiji, Tonga, Samoa, and Vanuatu eses are supported by substantial evidence (Flannery, 1995; Koopman, 1997). It is the (e.g., Tarburton, 2002). only microchiropteran in the Mariana Is- Emballonura s. rotensis once occurred lands, where the endemic subspecies E. s. on the five southernmost Mariana Islands of rotensis occurs (Yamashina, 1943; Lemke, Saipan, Tinian, Aguiguan, Rota, and Guam 1986; Flannery, 1995; Koopman, 1997). (Oustalet, 1895; Yamashina, 1943; Lemke, While some populations of E. semi- 1986; Koopman, 1997; Steadman, 1999; P. caudata have remained stable (e.g., Caro- Krutzsch, in litt.) and possibly on several is- lines — Bruner and Pratt, 1979; Wiles et al., lands in the northern portion of the archipel- 1997), many others have suffered dramatic ago (Lemke, 1986). Of these populations, Acta Chiropterologica, 6(2): 309–318, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

A test of the effects of climate and fruiting of Piper species (Piperaceae) on reproductive patterns of the bat Carollia perspicillata (Phyllostomidae)

MARCO AURELIO RIBEIRO MELLO1, GILBERTO MENEZES SCHITTINI2, PEDRO SELIG3, and HELENA GODOY BERGALLO4

1Departamento de Zoologia, Universidade Estadual de Campinas, CEP 13083-970, Cidade Universitária, Campinas, SP, Brasil; E-mail: [email protected] 2Secretaria de Políticas para o Desenvolvimento Sustentável, Ministério do Meio Ambiente, CEP 70068-900, Bloco B do 5º ao 9º andar, Brasilia, DF, Brasil 3HABTEC Engenharia Ambiental, CEP 20003-900, Av. 13 de Maio, nº 13. Centro, Rio de Janeiro, RJ, Brasil 4Departamento de Ecologia, Universidade do Estado do Rio de Janeiro, CEP 20599-900, Rua Sao Francisco Xavier 524, Maracana, Rio de Janeiro, RJ, Brasil

Bats of the genus Carollia feed mainly on plants of the genus Piper, dispersing their seeds in all localities where they occur together. Interactions among these bats, their food plants, and the climate were studied in an Atlantic Forest area in Southeastern Brazil. Path analysis was used to estimate the strength of direct and indirect effects through which variables determine the timing of bat reproduction. Temperature had a small direct influence, but a strong indirect one. Rainfall affected bat reproduction through indirect ways. Although the consumption of Piper fruits by bats did not have a significant influence, the timing of production of Piper fruits was a strong variable directly affecting bat reproduction. We therefore suggest that Piper plants and climate may play a keyrole in the timing of reproduction in C. perspicillata bats.

Key words: Atlantic Forest, frugivory, Carollia perspicillata, indirect effects, path analysis, reproduction, seasonality

INTRODUCTION Whittaker and Jones, 1994; Estrada and Coates-Estrada, 2001). Neotropical leaf-nosed bats (Chiroptera: Among the Phyllostomidae, bats of the Phyllostomidae) may be regarded as key- subfamily Carolliinae (genera Carollia and stone-species in forests (Marinho-Filho and Rhinophylla — Wetterer et al., 2000) inter- Sazima, 1998), because of their roles as act with plants mainly by fruit consumption seed dispersers and pollinators of many (Heithaus, 1982; Fleming, 1988). These bats plants (Kunz, 1982; Fleming, 1988; forage mainly at ground level in the under- Charles-Dominique, 1991; Nowak, 1994). story (Bernard, 2001), focusing on shrubs Consequently they are important to the and small trees, therefore being members of processes of ecological succession and the guild ‘highly cluttered space gleaning forest regeneration (Fenton et al., 1992; frugivores’ (Kalko, 1998). Acta Chiropterologica, 6(2): 319–334, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Bats of the Reserva Nacional Allpahuayo-Mishana, northeastern Peru, with notes on community structure

CHRISTINE L. HICE1, 2,PAUL M. VELAZCO3, 4, and MICHAEL R. WILLIG1

1Department of Biological Sciences and The Museum, Texas Tech University, Lubbock, TX 79409, USA 2Present address: Department of Biological Sciences, University of New Mexico, Albuquerque, NM 87131, USA; E-mail: [email protected] 3Division of , The Field Museum, Chicago, IL 60605, USA 4Department of Biological Sciences, University of Illinois at Chicago, Chicago, IL 60607, USA

An inventory of the bat fauna at the Reserva Nacional Allpahuayo-Mishana documented 63 species. Coupled with previous records, this becomes the second most species-rich site for bats in Peru, with 65 species. Reproductively, there was a peak in activity in the early rainy season (October–December) that steadily declined to a low in the late rainy or early dry season (May–June). The community was dominated by Carollia perspicillata in terms of relative abundance and biomass. Most species were frugivores (28) or insectivores (26), but frugivores were predominant in the community based on cumulative abundance and biomass. The bat fauna was sampled to 85–91% completion based on extrapolations of local species richness. Nonetheless, only 56% of the regional species pool was captured locally. The fauna was compositionally similar to that of other western Amazonian sites. Although the Reserva Nacional Allpahuayo-Mishana enjoys a relatively high protected status, this has yet to be translated into sustained conservation. As a hotspot of biodiversity in Peru, it deserves protection at the highest possible level.

Key words: Chiroptera, Peru, inventory, species richness, community structure, abundance, biomass, dominance, feeding guild, conservation, reproduction

INTRODUCTION World tropics, the richest bat fauna has been reported from Colombia, with 170 species The Chiroptera is the second most spe- (Rodríguez-Mahecha et al., 1995). How- cies-rich order of mammals after Rodentia, ever, twelve of those species are not repre- comprising 17 families, 177 genera, and sented by voucher specimens and their oc- 925 species (Koopman, 1993). Like most currences are inferred from distributional taxa, bat species richness peaks in records from bordering countries. As such, tropical regions (Findley, 1993; Willig et Peru ranks second in the Neotropics, with al., 2003) regardless of spatial scale (e.g., eight families, 61 genera, and 158 species Willig and Selcer, 1989; Willig and Sandlin, (Pacheco et al., 1995), of these, six species 1991; Willig and Lyons, 1998; Lyons and were documented recently, including Cen- Willig, 2002), with records of up to 175 tronycteris maximiliani (Hice and Solari, species from a single country (Indonesia — 2002), Micronycteris matses (Simmons et Mickleburgh et al., 2002). In the New al., 2002), Eumops maurus (Montambault, Acta Chiropterologica, 6(2): 335–346, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Detection and avoidance of harp traps by echolocating bats

NICHOLAS BERRY1, WILL O’CONNOR1, MARC W. HOLDERIED1, and GARETH JONES1, 2

1School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, United Kingdom 2Corresponding author: E-mail: [email protected]

Researchers often use harp traps and mist nets to capture bats, and need to be aware of factors that affect trap capture efficiency. Ultrasound reflects from small targets in a frequency-dependent manner, so we predicted that higher frequency sound pulses would return stronger echoes from the fine wires and net of bat traps than would lower frequency signals. We also predicted that mist nets would return stronger echoes than harp traps because mist nets are made of a higher density (and often diameter) of material. Ensonification experiments with pulses of 20–110 kHz showed that both harp traps and mist nets reflected higher frequency pulses more strongly. Pocketed areas of mist nets returned stronger echoes than harp traps although at some frequencies differences between trap types were small. Our results provide one verified reason why harp traps are more effective than mist nets at capturing bats, and also predict that bats using high frequency echolocation calls may be more difficult to trap than species emitting low frequency signals. Interspecies differences in how traps are detected are therefore likely sources of bias in field surveys. Observations of bats encountering harp traps in the field showed less than 4% of encounters resulted in capture, and only 8.8% of encounters could be interpreted as a failure to detect the trap. A comparison between two species that differ in echolocation call and flight characteristics (Rhinolophus hipposideros and Myotis nattereri) showed no difference in trap detection or avoidance. However, differences in behaviour during trap encounters were apparent.

Key words: echolocation, target strength, capture methods, survey bias

INTRODUCTION conspicuousness of mist nets may explain this disparity in capture rates. To date, the Researchers often need to capture echo- conspicuousness of bat traps as acoustic tar- locating bats to study and survey them. gets has not been studied, however. Capture methods are reviewed by Kunz Constantine (1958) developed the first and Kurta (1988), and two of the most fre- harp trap, which was later modified to quently employed capture methods are use increase effectiveness and portability (Tut- of harp traps (Kunz and Kurta, 1988; Mills tle, 1974; Tidemann and Woodside, 1978; et al., 1996; Duffy et al., 2000) and mist Francis, 1989). The design that is current- nets (e.g., Sedlock, 2001). Tidemann and ly most popular for use in temperate re- Woodside (1978) and Francis (1989) found gions makes use of two banks of vertical- harp traps much more effective than mist ly strung, 2.7–3.6 kg monofilament fish- nets at catching small rhinolophids and ing lines with spacings of 2.5 cm (Kunz and vespertilionids, and the greater acoustic Kurta, 1988). The trap is designed so that Acta Chiropterologica, 6(2): 347–363, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Data, sample sizes and statistics affect the recognition of species of bats by their echolocation calls

STEFANIA BISCARDI1, JAZMINE ORPRECIO1, M. BROCK FENTON1, 4, ASAF TSOAR2, and JOHN M. RATCLIFFE3

1Department of Biology, York University, North York, Ontario M3J 1P3, Canada 2Department of Evolution, Systematics and Ecology, The Hebrew University of Jerusalem, Jerusalem 91904, Israel 3Department of Zoology, University of Toronto at Mississauga, Mississauga, Ontario L5L 1C6, Canada 4Corresponding author: E-mail: [email protected]

Identification of bat species based on analysis of echolocation calls can be affected by the way data are manipulated, the diversity of species, and call variability. We document the effects of sample sizes and a priori assignment of calls by species on the outcome of discriminant function analysis (DFA) and multinomial logistic regression (MLR) of features of echolocation calls, and determine which features of calls are most useful for identification. We used recorded echolocation calls of eight species readily distinguishable by call features, including molossids, emballonurids and a moormopid recorded at sites in Belize, Brazil, and Mexico. On individual calls, we measured four features: frequency with most energy, highest and lowest frequencies and call durations obtained from sequences consisting of 10 calls. Cluster analysis and multiple analyses of variance indicated significant differences between the calls of different species. Outcomes of DFA and MLR were affected by both sample sizes (numbers of calls, numbers of sequences) and the subjective approach that researchers take to their data (i.e., categorizing calls or sequences of calls by species). Levels of variation in calls of some species in our sample often precluded the use of single calls in making call-based identifications. Accurate documentation of variability in echolocation behavior of sympatric bats is a prerequisite for an effective sound-based bat survey.

Key words: echolocation calls, emballonurids, molossids, mormoopids, variation, cluster analysis, discriminant function analysis, multiple logistic regression

INTRODUCTION to distinguish among species of bats by their echolocation calls, reflecting diagnostic pat- Recognizing species of echolocating terns of frequency change over time and bats by their calls has been valuable in species-specific frequencies in calls. The assessing the distribution of, and habitat feasibility of detecting and identifying a bat use by, bats not prone to capture (e.g., by its echolocation calls is a direct function Kalko, 1997; Ochoa et al., 2000). Many of call intensity. Species producing low in- authors (e.g., Fenton and Bell, 1981; Ah- tensity echolocation calls (the ‘whispering lén, 1990; O’Farrell and Miller, 1997, bats’ of Griffin, 1958) are more difficult to 1999; Russo and Jones, 2002; Rydell et al., sample than bats using high intensity echo- 2002) have demonstrated that it is possible location calls. Among species producing Acta Chiropterologica, 6(2): 373, 2004 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

REVIEWERS OF ARTICLES SUBMITTED TO VOLUME 6

The Editor is deeply grateful to the fol- Kearney, G. Kerth, T. Kokurewicz, T. H. lowing individuals, who contributed to the Kunz, A. Kurta, G. Kwiecinski, M. J. Lacki, reviewing process by submitting comments W. C. Lancaster, B. Law, B. Lim, C. López- on one or more manuscripts: González, G. Marimuthu, G. F. McCracken, R. A. Adams, L. F. Aguirre, J. Arroyo- R. A. Medellín, B. W. Miller, J. Molinari, R. Cabrales, R. M. R. Barclay, P. J. J. Bates, Mueller, R. D. Owen, V. Pacheco, S. Par- P. Benda, M. Brigham, H. Broders, P. sons, S. C. Pedersen, P. A. Racey, R. Ran- Charles-Dominique, G. Chaverri, M. Cie- some, M. Robinson, A. Rodríguez-Durán, chanowski, P. Cryan, G. Csorba, N. J. Czap- M. Ruedi, D. Russo, B. M. Siemers, S. M. lewski, V. Elangovan, J. Fahr, B. Fenton, C. Swift, P. Taylor, D. Thomas, C. Tidemann, Francis, P. W. Freeman, W. L. Gannon, C. M. D. Tuttle, V. van Cakenberghe, R. A. Van Hanlon, D. Happold, J. P. Hayes, A. Herrel, Den Bussche, J. P. Veilleux, P. M. Velazco, S. R. Hoofer, I. Horá…ek, J. M. Hutcheon, A. O. von Helversen, D. A. Waters, D. E. M. Hutson, C. Ibáñez, D. S. Jacobs, P. Wilson, J. Zima, A. Zubaid, and Y. Yom- Jenkins, G. Jones, J. Juste, E. K. V. Kalko, T. Tov.